IOSR Journal of Mechanical and Civil Engineering (IOSRJMCE)

ISSN: 2278-1684 Volume 1, Issue 6 (July-Aug 2012), PP 41-50

www.iosrjournals.org

Drag Reduction of V Shaped Ring Gutter of an Afterburner by

Tandem Bluff Bodies Using Cfd
S Srinivasan1, J J Isaac2, C.Rajashekar3, A.Arokkiaswamy4
1&4
Department of Aeronautical Engineering, MVJ College of Engineering, Visveswaraya Technological
University Channasandra, Near ITPB, Bangalore-67, Karnataka, India
234
National Aerospace LaboratoriesBangalore-17, Karnataka, India

Abstract: It is a well known approach that having bluff bodies in tandem will comparatively reduce the drag
generated by single bodies. Several experimental and numerical results in the past have proved this
phenomenon of drag reduction due to flow interference and patterns of vortex shedding. A V shaped gutter is
being used in afterburner of an aircraft gas turbine engine which serves to hold the flame when the afterburner
is switched ON. But when the burner is switched OFF the presence of gutter offers excess drag and total
pressure loss. Hence it is necessary to reduce the total pressure loss and drag generated by the gutter. In this
work an attempt has been made to understand the flow physics involved in keeping bodies in tandem and the
effect of change in drag coefficient. Several types of cross section of bodies are investigated both upstream and
downstream of the gutter and the drag coefficient is calculated. The flow simulation is done using CFD-ACE+,
commercially available CFD Software.
Key Words: Bluff bodies, Drag reduction, Tandem bodies, V gutter

I. INTRODUCTION:
Over the years, various research and development are carried out to improve the performance of
afterburner, both experimentally and theoretically. Computational methods have become highly useful tool to
design, develop, and analyze the performance of an afterburner with ease. The amount of drag generated by the
presence of flame stabilizer and the recirculation zones in the wake of the stabilizer are very important factor to
be considered for better performance of an afterburner. In the past, various research activities have been carried
out to reduce the drag of bluff bodies, by changing the orientation and arrangement of the bluff body with
respect to the flow such as, tandem, staggered, normal, and perpendicular to the flow. The above said researches
were carried out through experimental, theoretical and CFD methods. In this chapter a brief review of literature
closely related to the above mentioned studies are presented.
Nakanishi et al.. [1] Conducted experimental investigations on the effect of flame holder gutter shape
on afterburner performance. Conventional V-gutter has minimum drag and minimum pressure loss both in
burning and without burning condition and the flame holder shape had less effect on stability limits. In another
study by Kareem et al..[2] on cylinders in tandem, the flow in tandem arrangement in two regimes such as, for
spacing up to the critical spacing, the vortex street is suppressed behind the front cylinder and beyond this
critical spacing, both cylinders form vortex streets. Large values of drag co-efficient occur at ratio of cylinder
spacing to diameter of cylinder between 3 and 4. When the spacing is kept below the above mentioned value the
drag value is found to be less than the value obtained for the above mentioned spacing to diameter ratio. Similar
works has been carried out by various researchers to show the effect of flow interference and drag
characteristics’ when two bodies are in tandem.

II. PROBLEM STATEMENT

Computational Modelling of Flow over a V-gutter is done using CFD-ACE+ software. The flow
boundary conditions are given in the table 1. A steady, viscous flow model is considered and K-E turbulence
model is chosen. A convergence criterion of 10-6 is taken for the solution to converge. The gutter and bluff
bodies are considered as wall and the drag coefficient is calculated by summing up pressure and shear forces to
calculate the drag force and using the equation 1. The fig 1 shows the computational model chosen for the
analysis. The duct model surrounding the gutter is considered to be same in geometry as that of afterburner and
a cyclic symmetry boundary condition is applied to the side walls. The analysis is carried out for flow conditions
as shown in the table and for different configurations of bluff bodies in tandem as shown in figure.

www.iosrjournals.org 41 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD
TABLE I BOUNDARY CONDITIONS
Boundary Parameters Values
Inlet Mach number 0.35
Inlet Total Pressure, N/m2 225400
Inlet Total Temperature, K 1009.1
Inlet Air Mass flow, Kg/s 8.3
Inlet Velocity, m/s 220
Turbulence Level (%) 5
Turbulance kinetic energy,k 187.8
Dissipation rate,  655.57
Ratio of specific heat,  1.35
Exit Static Pressure, N/m2 224000

Cd = Fd / 0.5 ×  × V2× A---------------------------1

Where,
Cd - Coefficient of drag
Fd - Drag force in Newton
A - Projected area in m2
 - Density of the flow ahead of the gutter in Kg/m3
V - Velocity of the flow ahead of the gutter m/sec

Fig 1: Computational domain

III. RESULTS AND DISCUSSION

The flow physics over the gutter when it is single and when bluff bodies are in tandem upstream and
downstream are clearly captured and drag coefficient is estimated. Thirteen different configurations were
studied. The flow phenomenon clearly captured the recirculation zone in the wake of the gutter. Since the wake
width is also a measure of the drag coefficient, reduction in wake width brings a reduction in drag. The
configuration of bluff bodies in tandem to the gutter has shown diverse change in the drag coefficient.

3.1 CASE 1 - Bare gutter normal to the flow

CFD predictions have been carried out with bare gutter kept normal to the flow. The velocity variation
and flow pattern are shown in the figure. The drag force was found from the force summary output from the
CFD-ACE+ analysis software along the direction of flow over the gutter. The drag force and the coefficient of
drag are given below. It is seen from the results that, the coefficient of drag of the bare gutter is found to be 1.08
and the drag force (sum of pressure and viscous force) is found to be 286 Newton. The Fig 2 shown below
clearly shows the wake region formed behind the gutter and the pressure variation and flow separation. Further
analysis have been carried out for finding the possibility of reducing the Cd value below 1.08 of the bare gutter
by putting bodies in tandem to the gutter. The density and Velocity ahead of the gutter is constant for all the
configurations and their values are 0.76 kg/m3 and 130 m/sec respectively.

www.iosrjournals.org 42 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD

Fig 2: Flow pattern over the V-gutter

3.2 CASE 2 - Flat edged disc downstream of the gutter

Keeping the coefficient of drag of bare gutter as base, the flat edged disc of 35mm width is kept
downstream of the gutter and the coefficient of drag was calculated using the equation 1. The disc was aligned
axisymmetrically to the gutter and normal to the flow. The gap between the disc and gutter are varied as 10mm,
20mm and 30mm and the coefficient of drag was found to be 0.52, 0.53, and 0.55 respectively. The Cd of bare
gutter is 1.08 and hence a maximum of 31% of drag reduction has been achieved by keeping the disc in tandem.
The graph (Fig 3) shows the variation of Cd with spacing between the disc and the gutter. The reduction in drag
takes place by improved stream lining of the flow that is evident from the flow pattern shown in Fig 4. Since the
Cd is asymptotically increasing when the spacing is increased the analysis has been terminated with 30 mm gap.

Fig 3: Coefficient of drag Vs Gap between the flat edged disc and gutter

Fig 4: Flow pattern over the gutter with flat edged disc in tandem

3.3 CASE 3 - Rounded edged disc downstream of the gutter

The disc of 35mm width with its corners rounded is kept in tandem at spacings of 10mm, 20mm and
30mm and the coefficient of drag value has been predicted at these spacings. The difference between case 2 and
case 3 is change in the edges. It is found that the coefficient of drag with round edged disc has C d value less than
that of the flat edged disc. The Cd values at the spacings 10mm, 20mm, and 30mm are found to be, 0.29, 0.49
and 0.5, which is clear from the graph shown in Fig 5. This shows that rounding the corners will create less drag
than the body with sharp corners, which enhances a smooth flow over the bodies. Also the flow is smooth over
the bodies and wake length has been increased which is evident from Fig 6.Here also as the spacing is increased
the drag is increasing due to the flow interference.

www.iosrjournals.org 43 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD

Fig 5 Coefficient of drag Vs Gap between the rounded edged disc and gutter

Fig 6: Flow pattern of rounded edged disc downstream of the gutter

3.4 CASE 4 - Rounded edged disc of width 25mm downstream of the gutter
In case 4 analysis is carried out with reducing the width of the disk from 35mm to 25mm with rounded
edges and kept at spacings of 10mm, 20mm, 30mm and 40mm.The results obtained showed a notable thing that
the coefficient of drag was reducing up to 30mm and it increased rapidly for 40mm.The values of C d up to
30mm are 0.57, 0.54 and 0.52. For 40mm spacing the C d value has again raised and reached to 0.55.Thus the
critical spacing for this configuration is found as 30mm.After the critical spacing the bluffbody in tandem
behaves as a separate body and contributed higher drag force. In this configuration the body in tandem is well
kept in the wake of the gutter and hence the drag reduction has been achieved. Further a low-pressure region
exists in between the bodies causing reduction in base pressure.

Fig 7: Configuration and Cd Vs Spacing gap for case 4

3.5 CASE 5 - Tapered disc in downstream of the gutter

In case 5 instead of a disc of rounded edge and flat shapes, a tapered disc having the taper angle of 99
degrees is kept in tandem behind the gutter. Then the spacing G between them is varied for 10mm and 20mm. It
is predicted from the results that there is a reduction of C d of about 59% (i.e. Cd=0.44) at 10mm and after that it
has increased to a Cd value of 0.47 at 20mm.Since the Cd value was asymptotically increasing the analysis has
been done only up to 20mm. The variation of Cd with the spacing is given in Fig 8. Since the disc is tapered the
wake width is reduced and hence the Cd has reduced for the system.

www.iosrjournals.org 44 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD

Fig 8: Configuration and Cd Vs Spacing gap for case 5

3.6 CASE 6 - Slit of varying size made in 25 mm width disc downstream of the gutter
In this case, in the 25 mm width disc discussed in case 4, a hole or slit was made and it is increased in
size from 0 mm to 25mm i.e., with 25mm width disc to without disc configuration. Here only the slit size is
varied and the disc is kept at 30mm for which optimum value has reached in case 4. It is seen from the results of
case 4, that the coefficient of drag of gutter with disc of 25mm at 30mm spacing was 0.52. As the hole size is
increased in steps of 5mm at 10mm hole in the center of the disc, the Cd value has increased form 5mm hole size
to 25mm hole size. Hence providing the slit in the disc has only increased the Cd due to reduction in area though
the size of the wake is small, which is evident from equation 1. The variation of Cd is shown in Fig 9.

Fig 9: Configuration and Cd Vs hole size for case 6

3.7 CASE 7 - Round edged disc of 35mm width upstream of the gutter
In this case instead of having a body in downstream side of the gutter, because of its effect in
recirculation zone behind the gutter, a disc of 35mm width is kept upstream side of the gutter and analyzed
whether the Cd of the gutter can be reduced. The disc is also varied in spacing with respect to the gutter. From
the results it is observed that, even in the upstream side the value of coefficient of drag is found to be less than
that of the bare gutter for the spacings from 0mm to 20mm.The Cd values at various spacings shows a maximum
of 26% reduction in the coefficient of drag is there than the bare gutter. It is seen that at zero mm spacing the C d
is 0.79 and then it increased to 0.82 for 5mm gap and the critical spacing has occurred at 10mm where the C d
has come down to 0.81 and after that increased. After the critical spacing sudden shift in the flow takes place
and hence the drag has increased after critical spacing. Another reason is after that spacing the body will act as a
separate body and hence contribute to drag separately. From the flow pattern it is clear that the flow separating
from front body or disc again attaches to the rear end of the gutter there by not affecting the recirculation zone.
Also a low velocity and pressure zone exists behind the disc and the gutter. The variation of Cd with spacing is
shown in figure 10 and the flow pattern in figure 11

Fig 10: Configuration and Cd Vs Spacing gap for case 7

www.iosrjournals.org 45 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD

Fig 11: Flow pattern of the disc of 35mm width upstream of the gutter

3.8 CASE 8 - 20mm hole in the disc of 35mm width upstream of the gutter
In this case, the same disc taken in case 7 in the upstream side has been taken with a 20mm hole/slit in
the center of the disc and the distance between them was varied in steps of 5mm from 0mm to 20mm. The
results obtained shows that the coefficient of drag value has still increased than the plain disc. But it has dipped
to a Cd value of 0.96 at 15mm gap, which is the critical spacing for this configuration, because after 15mm C d is
again increasing. After 15mm, as said in case7 the disc acts as a separate body and hence there is an increase in
drag. The variation Cd values and flow pattern are shown in figure 12. It is very clearly seen that in all
configuration there exist a low pressure and low velocity zone behind the gutter enabling flame stabilization.

Fig 12: Configuration and Cd Vs Spacing gap for case 8

3.9 CASE 9 - 20mm hole in the disc of 35mm width downstream of the gutter
In this case, the same configuration discussed in case 8 is taken in the downstream side of the gutter to
predict the drag changes. Here also the spacing is varied from 0mm to 20mm in steps of 5mm.It is observed
from the results that the Cd value was less than the bare gutter and the percentage of reduction of drag is
maximum of 33.3 percent which is for the spacing of 0mm, which is practically difficult. Further if the spacing
is increased the Cd value increased than that of the disc without hole in the downstream side of the gutter as
discussed in case 3. This is due to the reduction in area, which is inversely proportional to the coefficient of drag
as said in equation 1. The length of the wake has also increased in this case. The C d variation with gap is shown
in figure 13.

Fig 13: Configuration and Cd Vs Spacing gap for case 9

3.10 CASE 10 - Two circular rings Upstream of the gutter

In this case the analysis has been carried out with two circular rings of 10mm diameter separated by a
distance of 15mm i.e., the width of the two rings is equal to the width of the gutter is kept upstream of the gutter.
The analysis is done in such a way that the two rings are moved with respect to the gutter from their position of
touching the gutter to a distance from the center of the ring to the center of the gutter of 20mm. The coefficient
of drag has been calculated for this configuration and from the results it is found that the Cd value was very high

www.iosrjournals.org 46 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD
compared to that of the Cd of the bare gutter. The variation of Cd with respect to spacing between them is shown
in figure 14.

Fig 14: Configuration and Cd Vs Spacing gap for case 10

3.11CASE 11: 35mm circular ring upstream of the gutter

In this case, the analysis is carried out for a circular ring of 30mm diameter upstream of the gutter in
various spacing. For this upstream configuration the coefficient of drag value has decreased to a maximum of 35
percentage than the bare gutter. From the flow pattern it is clearly seen that the flow separation is taking place in
a highly streamlined way. From the results it is observed that there is a reduction in drag as the spacing is
increased from 25mm to 35mm and again it has increased when the spacing is increased. This shows that the
25mm is the critical spacing for this case. The variation of Cd with respect to spacing is shown in figure 15.

Fig 15: Configuration and Cd Vs Spacing gap for case 11

3.12 CASE 12 - 20mm circular ring upstream of the gutter

In case 12, the diameter of the ring is reduced to 20mm compared to the earlier case where the diameter
is 35mm, and kept upstream of the gutter in various spacings. From the results obtained it is seen that the value
of Cd had reduced to a maximum of 25.9 percentage reduction than that of the bare gutter. From the flow pattern
shown in figure 17, it is seen that the flow separating from the upstream ring again joins in the tail end of the
gutter. Hence even the flow is stream lined the reduction in drag is not much. From the graph plotted for the
values obtained it is seen that for 25mm the coefficient of drag has reduced to 0.8 from 0.81 and again increased
to 0.82. Hence for this configuration the critical gap is 25mm after which the Cd increases asymptotically. Also
flow interference is happening in the gap between the gutter and the ring.

Fig 16: Configuration and Cd Vs Spacing gap for case 12

www.iosrjournals.org 47 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD

Fig 17: Flow pattern for 20mm circular ring upstream of the gutter

3.13 CASE 13 - 10mm circular ring upstream of the gutter

In this case, the analysis was carried out for the ring of 10mm diameter upstream of the gutter in
various spacings i.e., the diameter of the ring discussed in earlier cases is reduced to 10mm. In this analysis,
from the results obtained it is observed that the percentage of reduction is less as compared to earlier cases and it
is almost approaching the value of the bare gutter as the spacing is increased. The maximum amount of
reduction percentage is 15.7 percentage. From the flow pattern it is seen that the flow separating from the ring
again joins at the upstream side of the gutter and hence the reduction in Cd is very less. Also since the ring
diameter is 10mm which is less than the width of the gutter the flow over the gutter has changed and hence the
reduction percentage is less. From the graph it is clear that there is a sharp increase in the value of Cd at various
spacing. It is also clear that there is no dip in Cd value for this case. Hence for this case if the spacing is
increased the Cd value also increases. The variation of Cd with respect to various spacing is shown in figure 18.

Fig 18: Configuration and Cd Vs Spacing gap for case 13.

The following table II shows the summary of the results achieved for various configurations under various
orientations for the above discussed 13 cases.

TABLE II DRAG COEFFICIENT RESULT SUMMARY

Case
Configuration sketch Computed values of Cd
No

1 1.08

Spacing between bodies,

G (mm)
2 10 20 30
Coefficient of Drag, Cd
0.52 0.53 0.55

Spacing between bodies,

G (mm)
3 10 20 30
Coefficient of Drag, Cd
0.29 0.49 0.5

www.iosrjournals.org 48 | Page
Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD
Spacing between bodies, G
(mm)
4 10 20 30 40
Coefficient of Drag, Cd
0.57 0.54 0.52 0.55

Spacing between bodies,

G (mm)
5 10 20
Coefficient of Drag, Cd
0.44 0.47

Spacing between bodies, G (mm)

0 5 10 15 20 25
6
Coefficient of Drag, Cd
0.52 0.58 0.71 0.76 0.86 1.08

Spacing between bodies, G

(mm)
7 0 5 10 15 20
Coefficient of Drag, Cd
0.79 0.82 0.81 0.84 0.85

Spacing between bodies,

G (mm)
8 0 5 10 15 20
Coefficient of Drag, Cd
1.07 1.01 0.98 0.96 0.97
Spacing between bodies, G
(mm)
9 0 5 10 15 20
Coefficient of Drag, Cd
0.72 0.74 0.75 0.76 0.77

Spacing between bodies, G (mm)

10 Touched 5 10 15 20
Coefficient of Drag, Cd
1.24 1.33 1.12 1.14 1.17

Spacing between bodies,

G (mm)
11 25 35 45
Coefficient of Drag, Cd
0.71 0.7 0.76
Spacing between bodies,
G (mm)
12 20 25 30
Coefficient of Drag, Cd
0.81 0.8 0.82
Spacing between bodies,
G (mm)
13 15 20 25
Coefficient of Drag, Cd
0.91 0.92 0.94

www.iosrjournals.org 49 | Page
 Drag reduction of V Shaped Ring gutter of an afterburner by tandem bluff bodies using CFD
IV. CONCLUSION
It can be concluded from the CFD analysis that the drag of the V-gutter flame holder of an afterburner
can be reduced by providing a buff body in tandem either upstream or downstream. It is observed from the
studies that almost for all the case the drag of the gutter has reduced except case 10. The coefficient of drag is
comparatively less for the bodies having rounded corners than the bodies with flat edges as expected. Further,
the coefficient of drag varied with respect to the spacing between the bodies for all the cases. Further, from the
analysis it is evident that the drag reduction is more when the body is kept downstream of the gutter, where as it
is comparatively higher when it is kept upstream of the gutter. In the downstream side of the gutter, a maximum
of 73 percentage of drag reduction has been achieved for case 3 compared to that of the bare gutter and in the
upstream side, a maximum of 35 percentage of drag reduction has been achieved for case 11. Thus by keeping
the bluff bodies in tandem to the gutter with properly designed shape and size the drag of the gutter can be
reduced and hence the dry loss can be minimized thereby enhancing the performance of the afterburner.

V. ACKNOWLEDGEMENTS
The support rendered by National Aerospace Laboratories Propulsion division and MVJ College of
Engineering are kindly acknowledged.

References
[1] S.Nakanishi, W.W.Velie, and L.Bryant ,An investigation of effects of flame holder gutter shape on afterburner performance, NACA
Research memorandum, NACA RM E53J14, Feb-1954
[2] Kareem.A, Tracy Kijewski, Po-chien lu, Investigation of interference effects of group of finite cylinders, Journal of wind
engineering and Industrial aerodynamics, Vol-77-78, pp.503-520, Sep-1998.
[3] M.M.Zadravkovich, Review of flow interference between two cylinders in various arrangements, Journal of Fluids Engineering,
vol. 99, p. 618-633, Dec- 1977.
[4] T.Morel and M.Bohn, Flow over two circular discs in tandem, Journal fluids Engineering, Vol-102, pp.104-111, March-1980.
[5] P.F.Zhang, J.J.Wang, S.F.Lu.J.Mi, Aerodynamic characteristics of square cylinder with a rod in staggered arrangement,
Experiments in Fluids, Volume 38, Issue 4, pp.494-502, March-2005.
[6] Keith Koenig, Interference effects on the drag of bluff bodies in tandem, Phd project report submitted to California institute of
technology, URN No: etd-12042006-140552, May- 1978.
[7] S.V.Unaune and V.Ganesan, Analysis of reacting flows in an aero engine afterburner using computational fluid dynamics, Indian
Journal of Engineering and material sciences, Vol-11, pp.31-37, Feb-2004.
[8] V.Ganesan, Numerical computation of turbulent flow behind a conical baffle Indian journal of Technology, Vol-18, pp.447-450,
Nov-1980.
[9] A. Prasad and C.H.K. Williamson, A method for reduction of bluff body drag, Journal of wind engineering and Industrial
aerodynamics, Vol-69-71, pp.155-167, Jul-Oct-1997.
[10] David Bierman and William H.Herrnstein, The interference between struts in various combinations -NACA report no: 468,
Accession no: ADA278321, June-1933.
[11] Jing-Tang Yang and Go-Long Tsai, The wake flow structure of an open-slit V gutter, Experimental thermal and fluid science paper,
Vol-5, pp.685-696, 1992.
[12] Wolfgang Bergt, Aerodynamic flame holder, United states patent-3938325, Feb-1976.
[13] Robert.D.Blevins, Applied fluid dynamics hand book, Chapter 10-Fluidynamic Drag, Krieger Publishing company-1984.
[14] Mehdi R. Khorrami, David P.lockard, et.al, Simulation of bluff body flow interaction for noise source modeling, Report Number:
AIAA Paper 2006-3203.

www.iosrjournals.org 50 | Page